Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[SplitLogicalObjectFifos] Add support for dma tranposed on the target side #850

Open
wants to merge 1 commit into
base: main
Choose a base branch
from

Conversation

yzhang93
Copy link
Contributor

Previous PR #812 added support to transpose dma dimensions on the target side for control code optimization. However, this new dma addressing wasn't supported in SplitLogicalObjectFifosForConnectionReuse pass.

This PR keeps the original logic and most of the original codes while adding the support for the new dma format.

scf.forall (%arg4, %arg5) in (2, 2) {
%2 = affine.apply #map(%arg5)
%3 = affine.apply #map(%arg4)
%4 = amdaie.dma_cpy_nd(%0[0, 0, 0, 0] [2, 32, 2, 32] [2048, 32, 1024, 1], %1[%3, %2] [64, 64] [128, 1]) : (!amdaie.logicalobjectfifo<memref<2x2x32x32xi32, 1 : i32>>, !amdaie.logicalobjectfifo<memref<128x128xi32>>)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'd prefer adding an analysis between L2 ObjectFifo's shape <2x2x32x32> and the sizes involved in L2 as target [2, 32, 2, 32] to infer whether a transpose has taken place instead of using a pass flag for the same.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants